Designer ligands for beryllium

Article abstract of DOI:10.1021/ja047637t

We report the rational design of ligands that selectively bind beryllium. We selected two ligands to design Be based on binding polynulear species with a Be-O-Be motif: 2-hydroxyisophthalic acid (HIPA) and 2,3-dihydroxybenzoic acid (DHBA). All previous wo

Full text of DOI:10.1021/ja047637t

Published on Web 07/16/2004
Designer Ligands for Beryllium
Timothy S. Keizer, Nancy N. Sauer, and T. Mark McCleskey*
Chemistry DiVision (C-SIC, Mail Stop J514), Los Alamos National Laboratory, Los Alamos, New Mexico 87545
Received April 23, 2004; E-mail: tmark@lanl.gov
Despite the high toxicity of beryllium, it is widely used due to
its unique properties. Beryllium is toxic both as a carcinogen and
as the agent that initiates chronic beryllium disease (CBD). CBD,
a granulomatous lung disease, is a cell-mediated immune response
of inhaled beryllium in 6-20% of exposed individuals. The nature
and effects of CBD have been well-studied,1 but the role of
beryllium in triggering CBD is not well understood.2,3 The onset
of CBD can be delayed for 10-40 years after exposure, and there
is no clear dose response correlation. Research efforts tend to focus
on areas of biological4 and environmental5 effects of beryllium with
far less effort devoted to the speciation and interactions of beryllium
under physiological conditions. Currently, beryllium detection is
accomplished by ICP-AE (inductively coupled plasma-atomic
emission) or ICP-MS, requiring a large capital investment.5
A
selective fluorescent sensor would offer several advantages, includ-
ing: low-cost instrumentation, quick analysis, field portability, a
small sample size (100 µM) for analysis, and potential for imaging
in a biological system.
Figure 1. (Top) Structures of 2-hydroxyisophthalic acid (1) and 2,3-
dihydroxybenzoic acid (2). (Bottom) Two different binding pockets for Be
with HIPA (a) and DHBA (b).
Previous research has focused on the binding of beryllium by
ligands to make BeL and BeL2 species with chelating ligands such
as chromotropic acid.6-8 This strategy has resulted in ligands with
fairly high binding constants (chromotropic acid K ) 16.2) but poor
selectivity.6 Ligands such as chromotropic acid that show fluores-
cence changes upon binding beryllium are very intriguing for their
potential as fluorescent indicators, but the poor selectivity requires
the addition of ethylenediamine tetraacetic acid (EDTA) as a mask-
ing agent and limits their utility as a fluorescent sensor.9 Although
most literature has focused on BeL and BeL2 species, it is well-
known that the aqueous speciation of beryllium involves polynuclear
complexes such as Be3(OH)3.6 Recent reports have shown that
polynuclear species are also dominant in the speciation of Be
ring upon binding. The aryl ring provides the potential for
fluorescence, serves to add rigidity to the ligand’s binding site, and
increases the acidity of the ROH that serves as the bridging RO
unit in the Be-O-Be moiety. Binding of Be to HIPA and DHBA
ligands was characterized by a combination of 9Be and 13C NMR,
UV, and MS. By monitoring both 13C NMR to observe free/bound
9
ligand and Be NMR to observe free/bound Be as the Be:ligand
ratio was changed from 1:2 to 2:1, the binding ratio of beryllium
to the ligand was found to be 2:1. The 9Be NMRs of the 2:1
complex are consistent with the structures in Figure 1, where the
HIPA:Be complex has only one peak at 3.40 ppm for two identical
Be sites in a six-member ring and the DHBA:Be complex has two
peaks at 4.52 and 2.65 ppm corresponding to one Be in a five-
member ring and one Be in a six-member ring, respectively. The
UV/vis spectrum for Be2DHBA has a peak at 325 nm in agreement
with the fully deprotonated ligand in concentrated NaOH. The MS
for HIPA with Be shows a base ion at 429 m/z [Be4(HIPA)2(H2O)-
(OH)+] with a daughter peak at 215 [Be4(HIPA)2(H2O)22+, z ) 2],
suggesting that the species can readily dimerize as previously seen
for citric acid.10
The HIPA and DHBA ligands exhibit a red shift in the
absorbance of 10 and 14 nm, respectively, when Be is added in a
1:2 ratio of ligand to Be (Figure 2). The stability constants were
determined by spectrophotometric titrations (absorbance vs pH),
utilizing the data analysis program pHAB.14 The stability constants
for the Be/ligand complexes are summarized in Table 1. The simple
binding constants for BeL formation are slightly higher than that
of chromotropic acid (one of the highest reported binding constants
in the literature). The stability constants for the 1:2 species are
complexes of citric acid (CA) that feature a central Be-O-Be unit
with the O coming from the alkoxide in CA.10 The preferred geom-
etry for the Be in the Be-O-Be unit in CA analogues consists of
one five-member ring and one six-member ring. The Be-O-Be
motif with a bridging RO has been previously observed in crystal
t
structures for the Be glycolate system11 and with butoxide12 or
ethoxide13 as the bridging RO species, but in these cases the binding
constants for the ligands are very weak (log K1 Be-glycolate )
1.49).6 Herein, we report on beryllium binding to ligands that are
intentionally designed to bind polynuclear species of Be based on
the Be-O-Be motif. Two ligands, depicted in Figure 1, have been
evaluated: 2-hydroxyisophthalic acid (HIPA) and 2,3-dihydroxy-
benzoic acid (DHBA). Both ligands have a high affinity for
beryllium over a wide range of concentrations/pH and exhibit
luminescent changes upon binding the beryllium.
HIPA and DHBA were chosen to support the Be-O-Be motif
in two different manners; HIPA can form two six-member rings,
and DHBA can form one five-member ring and one six-member
extremely high and demonstrate the importance of the Be-O-Be
motif. Speciation plots based on the stability constants show that
9
9484
J. AM. CHEM. SOC. 2004, 126, 9484-9485
10.1021/ja047637t CCC: $27.50 © 2004 American Chemical Society

C O MMU N I C A T I O N S
Figure 2. (Left) Absorption spectrum of DHBA:Be (pH ) 7.5). (Right)
Absorption spectrum of HIPA:Be (pH ) 8.5). All solutions: [ligand] ) 1
x 10-4 M in 0.1 M KCl.
Figure 4. (Left) Emission spectrum of DHBA with Be at low concentration.
[Ligand] ) 10-5 M in HEPES buffer (0.05 M), pH ) 7, λexc ) 320 nm.
(Right) Fluorescence intensity as a function of [Be]. [Ligand] ) 10-5 M in
HEPES buffer (0.05 M), pH ) 7, λexc ) 320 nm.
Table 1. Stability Constants for Ligands 1 and 2 with Berylliuma
quotient
HIPA (1) log â
DHBA (2) log â
fluorescence remains minimal. The fluorescence intensity as a
function of [Be] (see Figure 4) gives a linear plot with no
interference from the presence of all seven metals down to 200
nM Be (1.8 ng/mL). This dramatic result demonstrates for the first
time that ligands can be rationally designed to selectively bind
beryllium based on binding polynuclear species. The DHBA ligand
detects Be at pH 7 with no interference, making it ideal for
biological imaging of Be.
[LH]/[L][H]
13.7
19.1
22.3
17.5
27.0
13.3
23.5
26.2
18.4
28.5
[LH2]/[LH][H]
[LH3]/[LH2][H]
[LBe]/[L][Be]
[LBe2]/[LBe][Be]
a All constants are determined at 298 K, 0.1 M KCl.
Although the importance of polynuclear species is known in
simple aqueous beryllium species with no ligands present, no
previous work has purposely exploited the potential of binding
polynuclear species for the detection of Be. The results with HIPA
and DHBA demonstrate that ligands can be designed to bind Be
strongly and selectively based on binding polynuclear structures.
The high binding constants observed for both ligands suggest that
binding polynuclear species may be a viable strategy for developing
therapeutic agents for CBD, and the stability of the bridging RO
group indicates that amino acids such as tyrosine, threonine, or
serine may play a role in biological interactions with Be clusters.
Future work will explore if such polynuclear speciation can offer
insight into physiological interactions and why Be leads to CBD
while there is no analogous chronic aluminum disease.
Figure 3. (Left) Emission spectra of DHBA with increasing [Be]. [Ligand]
) 10-5 M in HEPES buffer (0.05 M), pH ) 7, λexc ) 320 nm. (Right)
Emission spectra of HIPA with increasing [Be]. [Ligand] ) 10-5 M in
acetate buffer (0.05 M), pH ) 5, λexc ) 320 nm.
the Be2L species dominates even at concentrations as low as 1 µM.12
The Be2L species at low concentration (10 µM) is supported by
the presence of the Be4(HIPA)2(H2O)(OH)+ peak in the MS data.
As shown in Figure 3, both ligands demonstrate a strong
luminescence in the presence of Be when excited at 320 nm. The
HIPA ligand emits at 411 nm and shows a blue shift to 375 nm
upon addition of beryllium. The DHBA ligand has a nominal
emission spectrum, but in the presence of Be a strong emission is
observed at 402 nm. The fluorescence changes coupled with the
high binding constants of these ligands enable both ligands to be
used as fluorescent indicators for Be at very low levels. A 10 µM
solution of DHBA can be used to determine 50 nM of Be (4.5
pg/mL) as depicted in the emission spectrum in Figure 4. Other
fluorescent indicators have been reported for Be with similar
detection limits based on simple BeL binding, but all have severe
interferences from other metals such as aluminum and iron.5
Aluminum interferes strongly in most systems, because of the
similarity in the charge-to-size ratio and the fact that in the case of
simple metal-ligand binding with chelating ligands, the O-O
distance between the two nearest chelating oxygens in a tetra-
hedrally bound Be and an octahedrally bound Al are identical to
Acknowledgment. This work was supported by the Laboratory
Directed Research and Development program (LDRD) at Los
Alamos National Laboratory.
Supporting Information Available: Experimentals and a speciation
figure of the beryllium complexes (PDF). This material is available
free of charge via the Internet at http://pubs.acs.org.
References
(1) Saltini, C.; Amicosante, M. Am. J. Med. Sci. 2001, 321, 89-98.
(2) Eisenbud, M. Appl. Occup. EnViron. Hyg. 1998, 13, 25-31.
(3) Saltini, C.; Amicosante, M.; Franchi, A.; Lombardi, G.; Richeldi, L. Eur.
Respir. J. 1998, 12, 1463-1475.
(4) Rossman, M. D.; Preuss, O. P.; Powers, M. B. Beryllium: Biomedical
and EnVironmental Aspects; Williams and Wilkens: Baltimore, 1991.
(5) Taylor, T. P.; Ding, M.; Ehler, D. S.; Foreman, T. M.; Kaszuba, J. P.;
Sauer, N. N. J. EnVinon. Sci. Health 2003, A38, 439-469.
(6) Alderighi, L.; Gans, P.; Midollini, S.; Vacca, A. In AdVances in Inorganic
Chemistry: Main Group Chemistry; Sykes, A. G., Cowley, A. H., Eds.;
Academic Press: New York, 2000; Vol. 50, pp 109-172.
(7) Schmidbaur, H. Coord. Chem. ReV. 2001, 215, 223-242.
(8) Wong, C. Y.; Woolins, J. D. Coord. Chem. ReV. 1994, 130, 243-273.
(9) Matsumiya, H.; Hoshino, H.; Yotsuyanagi, T. Analyst 2001, 126, 2082-
2086.
(10) Keizer, T. S.; Sauer, N. N.; McCleskey, T. M. Inorg. Chem. In press.
(11) Kumberger, O.; Riede, J.; Schmidbaur, H. Z. Naturforsch., B: Chem. Sci.
1992, 47, 1717-1720.
within 0.02 Å. By designing a ligand to bind with the Be-O-Be
motif, we have imposed a tremendous selectivity for Be. We have
demonstrated this selectivity by looking at fluorescence as a function
of Be concentration in the presence of other metals. In a solution
with 10 µM of DHBA and a cocktail of metal ions including Al,
Fe, Cr, Cu, Zn, Cd, and Pb, all at 10 µM each, the background
(12) Bell, N. A.; Coates, G. E.; Shearer, H. M. M.; Twiss, J. Chem. Commun.
1983, 840.
(13) Dressel, M. P.; Nogai, S.; Berger, R. J. F.; Schmidbaur, H. Z. Naturforsch.,
B: Chem. Sci. 2003, 58, 173.
(14) Gans, P.; Sabatini, A.; Vacca, A. Ann. Chim. 1999, 89, 45-49.
JA047637T
9
J. AM. CHEM. SOC. VOL. 126, NO. 31, 2004 9485